M3N-VC: Multi-Modality Multi-Node Vehicle Classification

M3N-VC is a large-scale IoT vehicle monitoring dataset (18.26 hours), that consists of data collected in six different environments. We use 6 to 8 nodes in each environment to collect seismic and acoustic signals for multiple moving vehicles. The dataset supports a variety of research topics, including domain adaptation, multi-node pretraining, multi-node tracking, and vehicle classification, among others. For more details, pelase see readme.md. Further information is available at [1] and Github: https://github.com/restoreml/m3n-vc If you use M3N-VC dataset in your work, please cite: [1] Li, Jinyang, Yizhuo Chen, Ruijie Wang, Tomoyoshi Kimura, Tianshi Wang, You Lyu, Hongjue Zhao et al. "RestoreML: Practical Unsupervised Tuning of Deployed Intelligent IoT Systems." In 2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT), pp. 109-117. IEEE, 2025. @inproceedings{li2025restoreml, title={RestoreML: Practical unsupervised tuning of deployed intelligent iot systems}, author={Li, Jinyang and Chen, Yizhuo and Wang, Ruijie and Kimura, Tomoyoshi and Wang, Tianshi and Lyu, You and Zhao, Hongjue and Sun, Binqi and Wu, Shangchen and Hu, Yigong and others}, booktitle={2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT)}, pages={109--117}, year={2025}, organization={IEEE}}